An exec utive su m m ary for
m an ag ers and exec u tive
readers c an be foun d at the
end of th is artic le
Towards an institutional theory
of the dynamics of industrial
networks
Carlos Melo Brito
Lecturer, Faculty of Business, University of Porto, Porto, Portugal
Keywords Networking, Relationship marketing, Industrial relations, Organizational change
Abstract The study of interorganisationalcooperation has gained increased currency. An
important empirical and conceptual contribution in this field owes much to the network
approach. The picture provided by the network approach contrasts with other models that
regard cooperation as a mere contractual and legal inter-corporate connection. Whilst
accepting the existence of formal types of collaborative arrangements, the network
approach emphasises the importance of informal and emergent cooperation. This paper is
an attempt to extend the current perspective by focusing on interorganisational
cooperation in the context of collective action phenomena. These usually involve a large
number of actors concerned with the formulation of market rules, the prevention of
instability and disorder and, in general, the promotion or defence of their mutual
interests. The paper offers an institutional explanation of why and how collective actions
emerge and influence the shape and evolution of industrial networks.
There is nothing permanent except change (Heraclitus)
Econom ic exchange
process
Introduction
Change is a key feature of any economic system. Conventional theories have
traditionally conceived change and stability as contrasting characteristics. In
fact, change has been regarded the same as instability and stability the same as
absence of change. This view has, however, been challenged by the network
approach, a model developed by the International Marketing and Purchasing
(IMP) Group which regards change and stability as interwoven issues. On the
one hand, industrial networks are assumed to be stable inasmuch as the
economic exchange process tends to take place within the existing framework
of interorganisational relationships. In fact, ties and bonds amongst actors as
well as the cost of changing a web of relationships favour the establishment of
stable links. On the other hand, industrial networks can be regarded as ``living’’
structures in which the way actors, activities and resources relate to each other
is continuously changing not only because of the dynamics of the economic
process but also on account of movements of actors attempting to increase their
control over activities, resources and/or other actors.
Nonetheless, despite the existence of significant research directed towards
modelling the dynamic processes governing the evolution of industrial
networks, IMP researchers have so far mainly focused on processes of change
induced by economic and technological factors. However, as Douglass North,
the Nobel laureate, put it, there is a growing evidence that institutional
considerations are likely to be particularly relevant:
Institutions are the rules of the game in a society or, more formally, are the
humanly devised constraints that shape human interaction. In consequence they
structure incentives in human exchange, whether political, social, or economic.
Institutional change shapes the way societies evolve through time and hence is the
key to understanding historical change (North, 1991, p. 3).
The research register for this journal is available at
http://www.mcbup.com/research_registers
The current issue and full text archive of this journal is available at
http://www.emerald-library.com/ft
1 50
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 1 6 N O . 3 2 00 1, p p. 15 0-16 6, # M C B U N IV E R S IT Y P R E S S, 0 88 5-86 24
In this context, this paper attempts to extend current views on the dynamics
of industrial networks by focusing on the institutional role played by
collective action processes, i.e. movements of groups of actors aiming to
promote or defend their mutual interests. The paper begins with an overview
of some of the most insightful contributions in this field made by network
approach researchers. The second section addresses the so-called collective
action perspective. The section which follows, introducing the institutional
dimension of interorganizational relationships, develops a three-dimensional
model of industrial networks which is illustrated by a case study. The paper
ends with a number of recommendations for managers.
Process of interaction
Change in industrial networks
Firms are not independent entities acting on their own in the market. To
develop their activity they have to interact with other firms and
organisations, such as governmental departments, associations or regulatory
commissions. This process of interaction may give rise to lasting and stable
relations through which firms adjust products, production and routines. Such
relations are often built over a long period since their development requires
time and resources and may involve commitments for the future. Firms’
behaviour can thus be described as a cumulative process where relationships
are created and developed to guarantee firms’ control over the resources they
need, the selling of their output, and the pursuit of their objectives. This
system of interdependent organisations engaged in the production,
distribution and utilisation of goods and services, forms an industrial
network where the particular position assumed by one actor affects not only
its performance but also the evolution of other actors.
This vision owes much to the network approach, a model developed under
the stream of research carried out by IMP over the past 20 years (Hakansson,
1987; Axelsson and Easton, 1992; Naude and Turnbull, 1998). For a detailed
description of the history and provenance of the network approach, see
Turnbull et al. (1996) who trace the theory and development of the IMP
Group.
The struggle for control
Change is a key feature of any industrial network. According to Hakansson
(1992), it can be induced either by the dynamics of the economic cycle or by
actors struggling for control. The struggle for control is particularly
interesting in the context of this paper inasmuch as it involves two interwoven
and opposite tendencies. The first, named ``hierarchisation’’, leads to an
increased control of resources or activities by a decreasing number of actors.
The second, defined as ``extrication’’, occurs when such a control is
diminished and, as a consequence, is spread throughout the network.
The process can be described as follows. In industrial networks the most
general objective that actors pursue is to increase their power, which can be
achieved through an extended control over resources or activities. The
control process can be developed either directly (i.e. through ownership) or
indirectly (i.e. via relationships with other actors). In both cases, increasing
control requires resources. If control is expanded through ownership, actors
have to have the means (e.g. financial, material or human resources)
necessary to acquire new resources or the capacity to perform new activities.
If control is extended through relationships, actors have to get involved in
new webs of relations with other actors, which also requires resources.
Given the fact that resources are by definition limited, in both cases actors
have to decide how to allocate their resources. In other words, for each
situation they first have to decide how they wish to use their respective
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 51
resources to increase their power in the network. Following Hakansson
(1992), this means that gaining an increased control over resources or
activities will probably lead to a decreased control over other resources or
activities. When control is exerted indirectly, this is likely to be particularly
important in terms of pressure for change. Indeed, when an actor aims to
increase its power through an extended control, it has to develop new
relationships, in most cases at the expense of a reduction of its involvement
in other relationships. Therefore, the author contends, both a building/
organising process (hierarchisation) and a disrupting/disorganising process
(extrication) tend to occur. This means that for any individual actor, the
control process involves a set of different types of links with other actors. For
instance, it may encompass cooperative relationships among actors sharing
similar interests, but it may also give rise to conflicting relationships
involving actors with opposite interests.
Actors’ netw ork theories
As Hakansson and Johanson (1993) point out, in this process a key role is
likely to be played by the actors’ network theories, i.e. their perceptions
about the present relations between actors as well as their expectations and
intentions. Indeed, network theories have a bearing on the connections which
will lead to the process of change inasmuch as each actor’s network theory
tends not only to influence its actions and struggle for control, but can also
be communicated to other actors and thus influence their respective actions.
In general, changes induced by joint actions have much to do with these
processes. On the one hand, they tend to increase the power of the actors
involved in such actions so that they can expand their capacity to influence
the shape of the network(s) in which they are embedded. Moreover, as
recognised by Gadde and Hakansson (1992, p. 179), those changes are
``politically important’’ and must ``. . . be judged more in terms of their
effects on the power structure than on the way in which they influence the
activities or the use of resources from an economic point of view’’. In
addition, joint or collective actions constitute a means of aggregating,
aligning and mobilising dispersed interests so that changes can be pushed in
a particular direction. This concentration of disseminated and fragmented
power is thus an example of Hakansson’s concept of hierarchisation.
Continuous and
discontinuous industrial
networks
An important contribution to this issue is addressed by Lundgren (1992). The
author contends that two kinds of change may arise in industrial networks:
continuous and discontinuous. The former tends to result from the
coordination of activities undertaken by actors within the existing pattern of
exchange. From a network perspective, coordination involves both mutual
adaptation and learning which, influencing the resource structure, may create
the conditions for future changes within the network. On the other hand,
discontinuous changes generally occur as a result of mobilisation processes
which break old patterns of exchange and create new ones. As Lundgren
(1992) stresses, this perspective does not contradict the traditional thinking
in the network approach since the distinction between continuity and
discontinuity is a matter of degree and level of aggregation. While at higher
levels of aggregation almost all changes can be considered continuous in
nature, at lower levels discontinuities are more likely to occur and may
explain significant network changes at that level.
The process of mobilisation is discussed in detail by Lundgren (1992).
Quoting Scott (1987, p. 159), he defines mobilisation as ``. . . the process of
forming crowds, groups, associations and organisations for the pursuit of
collective goals’’. Although Lundgren (1992) states that a collective goal is
not a necessary condition for discontinuous changes, he recognises that the
1 52
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
mobilisation process is likely to evolve in a smoother way if actors have a
common vision, i.e. a common network theory, and share mutual objectives.
Moreover, the author separates the mobilisation process into two broad
categories. The first, named ``integrative mobilisation’’, refers to changes
based on existing activity cycles. The second, labelled ``changing
mobilisation’’, refers to changes induced by the creation of new activity
cycles and/or the breaking of some of the old ones. Both processes disturb
the existing patterns of coordination of activities and may give rise to new
networks or, at least, the fast re-structuring of the established network(s).
Nevertheless, Lundgren (1992) claims that interaction between different and
quite often contradictory factors makes things in practice much more
complex and difficult to anticipate. First, mobilisation is always constrained
by the existing resource and activity structures, which will probably slow the
progress of discontinuous change. Second, mobilisation requires a very
strong commitment from the actors involved. This means that it is more
likely to occur during unstable periods when problems are recognised by a
larger number of actors.
Moreover, a number of more recent works touch on these issues, albeit in a
superficial way. For instance, Hakansson and Snehota (1995, p. 273) state
that:
Changes aimed to stabilise or to change the networks are always a matter of two or
more actors working together with or against others. The actors adjust to others as
they know, from experience, that it is the only way to get others to adjust to them.
Interactions thus lead to joint actions among actors that shape the structure of
business networks and create the connected relationships and result in ties, links
and bonds . . . As a consequence no single actor alone is capable of maintaining or
changing the structure of the network.
And Hakansson and Henders (1995, p. 152) argue that:
Because the actors are connected through resource use in activities, the changes
implemented by one actor affect the change activities undertaken by other
actors. This interconnection, combined with the fact that many network
members must be mobilised in order to achieve a desired change, creates the
network dynamics.
Concentration of power
In conclusion, the network approach provides a conceptual basis for the
understanding of processes of change induced by joint or collective actions.
First, joint actions are in general concerned with processes of struggling for
control which are a major source of change in industrial networks. Second,
both Hakansson’s (1992) and Lundgren’s (1992) works are particularly
appropriate when analysing such type of phenomena. While Hakansson
stresses the importance of the concentration of power in fewer and fewer
actors, Lundgren calls the attention for discontinuous changes induced by
processes of mobilisation.
Nevertheless, the need for further improvements remains. In particular, it is
necessary to take into account the process through which collective forms of
organisation emerge, as well as the way they influence the shape of industrial
networks. The section which follows addresses these issues.
Collective action
The collective action perspective
In general, organisations which develop an activity for the provision of
collective benefits acquire the bulk of their resources from contributions
from their members. These contributions can take a variety of forms, such as
money, materials, services, time, or simply psychological commitment. Once
control over such resources is achieved, the organisation may use them to
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 53
attain two main collective goals: to produce a direct product or service to its
membership; and/or to influence other actors’ activities. Collective action
problems may arise when members of such organisations are free to choose
whether to contribute or not to the provision of the collective benefit. In a
situation like this, self-interest may induce people not to join the action
because they may benefit from the effort of the others without paying for it.
In this context, free-riding is the opposite of cooperation since this means to
contribute to the collective action and, thus, to relegate self-interest to second
place. In short, cooperation becomes a matter of a tension between individual
and collective interests which, in most cases, assumes the nature of a conflict
between short- and long-term interests.
Individual versus collective
rationality
Most of the theoretical and empirical research on problems of individual
versus collective rationality owes much to Olson’s (1965) seminal
contribution addressed in The Logic of Collective Action. This book,
beginning with an analysis and conceptualisation of the so-called collective
action problem, attempts to establish ways of inducing people to cooperate
on a collective basis. Olson (1965, p. 14) concentrates his efforts on the study
of organisations which provide public or collective goods, i.e. ``goods that
are available to everyone if they are available to anyone’’. Such
organisations may assume, for example, the form of business cartels,
professional pressure groups, trade unions, and civil rights groups. In this
type of organisational structure, Olson (1965) observes that collective
benefits are, in most cases, insufficient for motivating individual
contributions. In fact, when large collectivities offer public goods as their
sole incentive both the ``imperceptible effect’’ and the ``free-rider problem’’
are likely to jeopardise collective actions. First, as groups become larger,
individual contributions to the collective action tend to be greater than the
perceived individual proportion of the public good shared by each member.
In other words, it becomes increasingly difficult for each member to
ascertain what the returns are on his or her contribution. Second, in large
groups members are more likely to tolerate instances of non-contribution. In
this context, any member acting on a rational and utilitarian basis may
maximise his or her benefits by not making any effort on behalf of the group.
If a significant number of members adopt this pattern of behaviour, then
suboptimal amounts of the collective good will be produced and, at the
extreme, the collective action carried out by the group will be nil.
Extending his analysis to small groups, Olson (1965) states that these are
more likely to mobilise interests and, therefore, to induce individual
contributions to collective actions. The author suggests three explanations
for this. First, the ``imperceptible effect’’ becomes less relevant. The smaller
the group is, the more each actor can anticipate and perceive his or her
benefit from the collective action. Second, in small groups the ``free-rider
problem’’ is less likely to occur because members tend to react if one of them
does not contribute to the provision of the collective good. Finally, the
smaller the group is, the lower organisation costs tend to be, e.g. the costs of
providing the collective good, the costs of maintaining the organisation, and
the costs of communication and bargaining among group members.
‘‘By-product’’ theory
1 54
Nevertheless, even in small groups the initial problem remains: how to
motivate group members to contribute to the collective action? Olson
contends that this problem can only be solved if individual rewards are
offered to complement the collective benefit. According to his ``by-product’’
theory, these selective and private incentives are likely to play a key role in
motivating individual contributions. Such incentives can be either positive or
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
negative, and are limited to two types: monetary and social incentives. While
the former is mainly economic in nature, the latter is essentially related to
each actor’s desire for approbation and the dislike of disapprobation. Social
incentives work especially through the mechanism of criticism and shunning
by the other members of the group.
Collective action issues and
problem s
Much of the theoretical and empirical research on collective action issues
can be traced back to Olson’s (1965) seminal and very influential work. His
legacy and the concern with collective action problems have become
widespread across a broad range of disciplines. UdeÂhn (1993) provides a
comprehensive overview of Olson’s legacy and Heckathorn’s (1996)
analysis attempts to supply an integrative framework to examine collective
action problems.
Nonetheless, Olson’s (1965) contribution has been subject to numerous
criticisms. Firstly, his formulation of the collective action problem as
essentially an N-person prisoner’s dilemma (Hardin, 1982) has been
considerably refined and elaborated upon. As some commentators have
noted, self-interest alone is unlikely to account for the emergence of
collective actions (Miller, 1992). Secondly, Olson’s (1965) theory is
guilty of what Macneil (1990) terms the Hobbesian fallacy: modelling
man as an individual and independent atom rather than a social actor
within an interdependent community of other actors. Finally, Olson’s
formulation of the problem fails to explain the emergence of institutional
groups of interests (Van Waarden, 1992) and social movements (Munck,
1995).
Selective incentives
Of the subsequent refinements to Olson’s approach, two deserve special
attention in the context of this paper. The first relates to Taylor and
Singleton’s (1993) and Posner’s (1996) attempts to extend Olson’s (1965)
notion of selective incentives to encompass notions such as ``community
spirit’’ or ``group solidarity’’, and explain why in some cases groups are able
to solve collective problems by themselves and in others the problem is only
solved by recourse to an external agent (e.g. the state). Whereas Taylor and
Singleton (1993) are concerned with the conditions that enable endogenous
resolution of collective action problems, Posner (1996) is concerned with the
influence that interaction between legal and non-legal sanctions has on the
behaviour of collective forms of organisation.
Hom ogeneity and
interchangeability of
individuals
The second useful extension of Olson’s (1965) work relates to the
relaxation of the assumptions concerning homogeneity and
interchangeability of individuals involved in collective actions (UdeÂhn,
1993). Whereas Olson (1965) stresses the role of political entrepreneurs in
large and heterogeneous groups, often a small subset of interested actors ±
i.e. the ``critical mass’’, according to Oliver and Marwell (1988)
terminology ± may be sufficient to mobilise time, money and other
resources to produce a collective benefit despite the majority of members
contribute little or nothing.
In concludsion, research on the collective action phenomenon encompasses
some key issues that can effectively contribute to the construction of a model
of change in industrial networks induced by mobilisation of interests. The
conceptual framework developed around the collective action problem is
undoubtedly relevant since the tension between individual and collective
interests is a key element for understanding the emergence (or absence) of
collective forms of organisation as well as their impact on the shape of
industrial networks.
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 55
Vertical and horizontal
relationships
The institutional driving force
The network approach has traditionally been considered a flat, twodimensional model. On the one hand, it encompasses vertical relationships
along the manufacturing and distribution channel (Ford et al., 1998). The
buyer-seller relationship has historically been the first to be studied, but
other links have also been considered, e.g. between suppliers and
customers of customers. The second dimension, referred to as horizontal,
addresses relationships among competitors (ArauÂjo and Mouzas, 1997).
Taking into account the role played by collective actions, a new type of
actor has to be considered: the ``collective actor’’. From an
interorganisational point of view, a collective actor is a net of
relationships created in order to cope with a collectively perceived and
shared issue. In other words, a collective actor is an issue-based net (Brito,
1999).
Collective actors may or may not adopt formalised structures. The former are
those that, being created through an explicit contract, assume a formal
structure. They encompass such different forms as trade associations, farmer
cooperatives, consortia of firms for joint sourcing, and regulatory
commissions. However, non-formalised collective actors may also come into
existence as simply virtual nets of relationships without any kind of formal
organisational structure supporting them. This may be the case of informal
groups of actors developing lobbying activities. They exist as a collective
actor since a net of relationships has been set and developed in order to
perform a particular collective action. In short, formalised or not, collective
actors consist of nets of relationships that support the formulation of internal
rules, the making of decisions, and the implementation and execution of their
actions.
Collective actors
A collective actor becomes stronger to the extent that it is able to firmly
associate, align and move in the same direction a large number of different
elements, and it can intervene, stabilise or change the linkages and
associations that constitute the network’s structures. The strength of a
collective actor depends on its capacity to collectively interpret issues, to
align interests, and to intervene or, quite often, to interrupt others’
interventions. Nevertheless, collective actors may not survive for long
periods of time, especially if they have an informal nature. Aggregating and
aligning disparate sets of interests into a coherent set of objectives and
programmes of action as well as promoting cooperation amongst selfish and
potentially opportunistic agents is an arduous task. Priorities and
circumstances change and collective actors that do not fulfil early promises
cannot hope to survive for long. They constitute vehicles for aggregating,
aligning and mobilising interests as well as pushing change in a particular
direction for a period of time. But often they are relatively fragile, unable to
resist trials of strength or by-pass obstacles placed in their way without
loosing significant momentum and credibility. Moreover, collective actors
are sometimes set up simply to address the resolution of a specific problem
and as soon as progress is made on that front, their existence is no longer
justified.
Collective actors affect not only the organisation of individual actors and
their strategies, but also the structure of relationships and the balance of
power within industrial networks. In this sense, cooperation, assuming the
form of a collective action, tends to play a key role in shaping the ``rules of
the game’’ and the structure of the network. From this point of view,
cooperation, complementarity and coordination must be perceived in the
1 56
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
context of groups of actors (i.e. nets), rather than merely at the dyadic level
as Mattsson (1985) suggested.
Institutional relationship
Furthermore, the existence of collective actors and the relationships
established around them introduces a new relational dimension to the
traditional network approach model: the ``institutional relationship’’. In this
context, the model assumes a three-dimensional nature.
Figure 1 shows the economic network which corresponds to the twodimensional industrial network as it has traditionally been addressed by the
network approach. Above it, collective actors emerge as a result of a process
of institutional aggregation of a range of dispersed and fragmented interests
into an expected coherent and unified action. In this sense, they are nets of
relationships created in order to cope with a particular issue (or set of issues)
concerning a group of actors. This means that the immediate objective of a
collective actor is to provide a product or service to its membership, and/or
to directly influence other actors’ activities. In any case, given the
interdependence among actors (collective or not), any joint action will
always affect other actors. Accordingly, irrespective of the specific
objectives pursued, the ultimate goal of a collective actor is to reinforce its
members’ power within the overall network. This is achieved through an
increased control over resources, activities and/or other actors so that they
can influence the structure and future evolution of the system. Although such
a goal may quite often not be clearly perceived as such, it is likely to be
achieved through the formulation or reformulation of the rule system that
guides and regulates business practice, the prevention of instability and
disorder, and the support to common values and beliefs.
Equilibria and disequilibria
According to this model, network changes are likely to be the product of a
series of equilibria and disequilibria ± most often precarious in nature ±
which result from the interdependence among three components of the
system:
(1) the distribution of power between individual actors;
(2) the distribution of power between their respective representative bodies;
(3) and the interaction process involving both individual and collective
actors.
Figure 1. The three-dimensional model of industrial networks
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 57
In these circumstances, equilibrium and disequilibrium become a matter of a
broad process of interaction which involves individual relationships as well
as multiple forms of institutional relations within, across and among
collectivities which aggregate different types of actors sharing similar
interests concerned with the multiplicity of issues that are likely to affect the
industrial network.
In sum, institutional relationships must be regarded both as network
structuring tolls and sources of change. They are network structuring tools
inasmuch as they affect the processes and structure of industrial networks.
They are sources of change because they are likely to give rise to imbalances
that, being the outcome of a process of equilibria and disequilibria, tend to
permanently create the conditions for alteration, transformation and movement.
Research project
A case study: the port wine industry
The model described before was developed on the basis of a research project
focusing on the port wine industry. Port is a fortified wine named after Porto,
the second largest Portuguese city from where it has traditionally been
shipped. However, port begins its life in the Douro valley, a region which
starts some 100km east of Porto and extends as far as the Spanish border.
This is an administratively demarcated region where an estimated 30,000
farmers grow what is considered to be one of the great wines in the world.
Despite the intervention of tens of thousands of actors in port production and
shipping, they can be grouped into four main categories (Figure 2):
(1) farmers,
(2) wine cooperative societies,
(3) shippers and
(4) distributors.
Traditionally, grape growing and port making were undertaken by
independent farmers who sold their wine to the shipping-houses. These, in
turn, organised the transport of the wine from the valley down the river
Douro, and stored and aged it in their lodges located in the Porto suburb of
Vila Nova de Gaia. Furthermore, the shippers promoted and sold the wine all
over the world to hundreds of distributors.
Vertical (dis)integration
Over the past decades some meaningful changes have occurred in this
process, most of them related to the emergence of new actors and processes
of vertical (dis)integration. First, a number of wine cooperatives in the Douro
region were created in the early 1950s. Originally, these co-ops received the
Figure 2. The port wine supply chain
1 58
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
grapes from the farmers, produced the wine and then sold it to the shippers.
Today, some cooperatives are also selling part of their production directly to
the distributors. Second, most farmers, who continued to sell their production
to the shippers, have closed their wineries. This means that they are
exclusively producing grapes while a major part of the wine is made by the
shippers. This shift has been further reinforced because most shippers have
also acquired large estates (called quintas) in the Douro valley where they
grow some of the grapes they need. Finally, since the mid-1980s, some
farmers have began to age and bottle their own wine, and ship it directly
from the Douro valley.
Production and distribution
The port wine industry is characterised by a number of aspects that make the
study of change induced by institutional driving forces particularly
interesting. First, it is embedded in a stable, mature and differentiated social
structure. The split between production and distribution is both social and
geographic. As stated before, all but the final phase of production is centred
in the farms of the Douro valley, whereas Porto and Vila Nova de Gaia are
the distribution and trading centres. Socially, the split is between a rural,
provincial and relatively isolated social structure dedicated to grape
production, and a more cosmopolitan, wealthier and better educated eÂlite
focusing on the technical aspects of production and trade. Second, the clear
geographical boundedness of the network and its dependence on activities
directly tied to the land make it relatively easy, in practice, to identify the
actors involved despite the existence of several tens of thousands of agents
concerned with both production and trade.
Technological change
Moreover, in the port industry, technological change relating to viticulture
and vinification techniques is relatively less important than other forms of
change such as the ones caused by market conditions. Finally, it is an
industry where power asymmetries (concerning both individual and
collective actors) play a crucial role in, for example, economic exchange
relationships, the shaping of actors’ perceptions and beliefs, and the
processes of network structuring.
Figure 3 shows the importance of the institutional dimension within the port
wine network. It makes clear the existence of two basic nets of relationships.
One is connected with the production of port in the Douro valley. It includes
Figure 3. The port wine network
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 59
not only the farmers and wine cooperatives but also a number of associations
such as the Casa do Douro (the Farmers’ and Growers’ Federation), the
AssociacË aÄo dos Produtores Engarrafadores (the Port Wine Growers’ and
Bottlers’ Association), and the UniaÄo das Adegas Cooperativas (the Wine
Cooperative Societies’ Association). The other net is centred around the
trade of port. It includes the shippers, Gruporto (a consortium of shippinghouses) as well as a number of associations, e.g. the AssociacËaÄo das
Empresas de Vinho do Porto (the Port Wine Shippers’ Trade Association)
and the British Association. The Douro valley net and the trade net are
connected by both economic and non-economic links. The former essentially
involves the farmer-shipper relationship. The latter encompasses the
relationships established with and through the Instituto do Vinho do Porto
(the Port Wine Institute, the governmental organisation which controls and
supervises the production and trade of port) as well as with and through other
collectivities such as Gruporto.
Industrial network
The port wine industrial network is thus characterised by an important
feature which cannot be neglected: the existence of a significant number of
collective actors such as the Farmers’ and Growers’ Federation, the Port
Wine Shippers’ Trade Association and Gruporto. Most of them are attempts
to influence the shape of the network inasmuch as they contribute to the
creation of rule systems that guide and regulate business practices, uphold
common values and, in general, tend to reinforce the strategic position of
their members in the network.
For instance, ADVID is an organisation made up of both farmers and
shippers whose members cooperate in terms of technical research on
viticulture and vinification. The outcome of the research is publicised by
books, seminars and conferences in order to disseminate information and
knowledge to other actors in the sector. The Port Wine Brotherhood is
another collective actor. Created in the 1980s by managers and owners of
shipping-houses, it aims the diffusion, promotion and consolidation of the
worldwide reputation of port wine. Acting as guardian of traditions, the
Brotherhood undertakes a number of initiatives such as solemn
enthronements of new members and blind vintage tasting. The Port Wine
Growers’ and Bottlers’ Association is also a collective actor. It was created
by a number of well-known producers aiming to promote and increase the
exports of port directly from the Douro valley. In short, these collectivities
concern themselves with defining, advancing and promoting their members’
interests across a broad spectrum of issues as well as attempting to prevent
instability and disorder in their dealings with other groups of actors.
Econom ic and institutional
relationships
In conclusion, different types of actors in the port network use different forms
of both economic and institutional relationships in their efforts to rewrite the
scripts of interaction, change (or preserve) the rules that govern business and
social practice, and reverse balances of power. The motives and dynamics
underlying the formation and persistence of these coalitions of actors as well
as their role and impact on the evolution of the port industrial network are
critical to the understanding of change in this particular business system.
Managerial implications
The importance of collective actions aggregating economic, social and
political interests is not an idiosyncrasy of the port wine industry since it is
also a typical feature of many other industrial networks such as the regions of
Baden-Wurttemberg in Germany, Jutland in Denmark, Emilia-Romagna in
Italy, Smaland in Sweden, and Oyonnax in France. By way of illustration, let
1 60
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
us take the case of the south-west German industrial district of BadenWurttemberg. As Sabel et al. (1987) claim, this is an economic system where
small and medium-sized firms are linked with and through a wide range of
collective forms of organisation that affect the fortunes of the industry as a
whole.
This issue deserves the attention of Herrigel (1993) who argues that the
dynamics of this economic system is mainly achieved through a collective
effort:
The collective process of negotiation over system reform is organised through
formal and informal networks. Industrialists, bankers, trade association officials,
and government people hold official positions on the governing boards and
advisory councils of the Fachhochschulen, banks, small and medium-sized firms,
and industry associations (Herrigel, 1993, p. 232).
Cooperative relationships
in industrial districts
A similar situation is found in the Third Italy which encompasses a number
of industrial districts located in the provinces of Abruzzi, Emilia-Romagna,
the Marche, Tuscany and Veneto. Best (1990, p. 238), in his book The New
Competition, states that:
The Third Italy is a goldmine for studying institutions by which individuals can
achieve the benefits for joint action that are beyond the reach of individual action.
The collective character of most of the cooperative relationships in industrial
districts is summarised by Powell (1990, p. 322):
. . . non-market, non-hierarchical modes of exchange represent a particular form of
collective action, one in which cooperation can be sustained over the long run as
an effective arrangement.
Regionally-based
econom ies
This shows that the nature and the importance of collective action within
industrial districts is similar to the one found in the port wine industry.
However, we lack the relevant yardstick to compare the importance of
collective actions across different industrial districts and, although collective
action is frequently mentioned as playing an important role in the operation
of industrial districts, I know of no empirical study which has focused
specifically on this phenomenon. From my point of view, the key conclusion
is that, in all regionally-based economies, collective action movements seem
to play a significant role in the shaping the structure and processes of the
industrial networks in which they are embedded. As Herrigel (1993, p. 227)
puts it, referring to the case of Baden-Wurttemberg, industrial power refers
to ``. . . the capacity to participate in the (re)definition and (re)composition of
the organisational and institutional structure of the industrial system itself’’
rather than on the mere ``. . . capacity of individual units to control and/or
distribute resources within a given system of production’’. Herrigel (1993,
p. 227) makes evident the collective nature of such movements by arguing
that ``. . . power is more of matter of identity and collective understanding
than one of resources and control’’. And Lorenz (1992, p. 195) adds:
Cooperation among producers in industrial districts has two principal aspects. It
takes the form of the provision of collective goods . . . Collective goods generally
are provided through the auspices of some local institution: business association,
trade union, or possibly municipal or regional government. Cooperation also takes
the form of adherence by producers to a set of norms of competition.
In conclusion, there are two important similarities between the port wine
industry and industrial districts. The first has to do with their basic
characteristics, i.e. clear geographical boundaries, the industrial population
being made up of many small and medium-sized economic units, and so
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 61
forth. The second is more fundamental and has to do with the nature of the
collective action phenomenon. In both cases it cannot be seen simply in
terms of the provision of a collective good. Rather, collective action is a
vehicle for changing the structure and interaction rules of the industrial
network.
In this context, the managerial implications of the model are valuable,
especially regarding the mangers’ scope of action and the role of interests.
M anaging relationships
Managers’ scope of action
Firms are not atomistic units acting on their own in a faceless environment.
To develop their activity, they have to interact with each other. This tends to
give rise to close and stable relationships through which firms adjust
resources and activities. Managers’ action can thus be described as a process
of managing relationships in order to guarantee the control over resources,
activities and even other actors so that his/her firm can achieve its objectives.
This perspective leads us to regard management as the art of relating firms
with the environment rather than a mere adaptation to the environment.
Therefore, the collective action phenomenon and the institutional dimension
of relationships extend the scope of managers’ action. Each firm’s evolution
depends both on its actual position in the network, and on the conditions
stemming from the structure and processes of the overall network. In this
context, firms may, eventually, influence the shape of the industrial network
in which they are embedded through collective actions carried out with other
actors sharing similar interests. In this sense, collective actions assume a
pre-competitive character.
Social and political
networks
The role of interests
Collective actions are undertaken by groups of actors often connected within
and across networks qualitatively different in nature. This means that if
financial statements, technological factors, product lines, human resources,
and so forth, are undoubtedly important, managers cannot forget that firms
are embedded not only in economic but also in social and political networks
whose impact cannot be neglected.
In short, managers must be aware that interests may be influenced by
economic reasons, but these are not the exclusive sine qua non of interests.
In other words, business is not only driven by economic reasons, but also by
interests which often assume an economic, social and political nature.
Conclusion
One of the characteristics of the model described in this paper is that
interorganisational relations are often polymorphic and fluid over time. This
inconsistency is likely to allow for the reconfiguration of nets of
relationships according to changing constraints and/or demands. In these
circumstances, industrial networks become interaction fields characterised
by an increasing degree of polyvalence and volatility where issues
crystallised around common and conflicting interests are likely to assume a
crucial role in explaining actors’ behaviours and networks processes. The
model proposed aims to deal with these new types of collective issues which
tend to give rise to domain-oriented collective actors, cross-cutting the
established order of economic exchange relationships.
This conceptual development, extending the scope of applicability of the
industrial networks model, opens up new avenues for empirical studies using
the network approach. By widening the concept of exchange to include
1 62
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01
forms of exchange other than straightforward economic exchange, and by
developing and refining the concept of ``actor’’, a plethora of new
possibilities emerge for network empirical studies. By way of example:
collective strategic positions and collective strategic actions in industrial
networks; multi-level relationships between different classes of actors; the
role of regulatory bodies in industrial networks; the emergence of industry/
trade associations as well as technical and product standards bodies; and the
formation and development of industrial districts and geographically
concentrated industries from a network perspective.
References
ArauÂjo, L. and Mouzas, S. (1997), ``Competition and cooperation in vertical marketing
systems’’, in Gemunden, G., Ritter, T. and Walter, A. (Eds), Relationships and Networks
in International Markets, Elsevier Science, Oxford.
Axelsson, B. and Easton, G. (Eds) (1992), Industrial Networks: A New View of Reality,
Routledge, London.
Best, M. (1990), The New Competition, Polity Press, Cambridge.
Brito, C. (1999), ``Issue-based nets: a methodological approach to the sampling issue in
industrial networks research’’, Qualitative Market Research: An International Journal,
Vol. 2 No. 2, pp. 92-102.
Ford, D., Gadde, L.-E., Hakansson, H., Lundgren, A., Snehota, I., Turnbull, P. and Wilson, D.
(1998), Managing Business Relationships, John Wiley & Sons, Chichester.
Gadde, L.-E. and Hakansson, H. (1992), ``Analysing change and stability in distribution
channels ± a network approach’’, in Axelsson, B. and Easton, G. (Eds), Industrial
Networks: A New View of Reality, Routledge, London.
Hakansson, H. (Ed.) (1987), Industrial Technological Development: A Network Approach,
Croom Helm, London.
Hakansson, H. (1992), ``Evolution processes in industrial networks’’, in Axelsson, B. and
Easton, G. (Eds), Industrial Networks: A New View of Reality, Routledge, London.
Hakansson, H. and Henders, B. (1995), ``Networks dynamics: forces and processes underlying
evolution and revolution in business networks’’, in Moller, K. and Wilson, D. (Eds),
Business Marketing: An Interaction and Network Perspective, Kluwer Academic
Publishers, Boston, MA.
Hakansson, H. and Johanson, J. (1993), ``The network as a governance structure: interfirm
cooperation beyond markets and hierarchies’’, in Grabher, G. (Ed.), The Embedded Firm:
On the Socioeconomics of Industrial Networks, Routledge, London.
Hakansson, H. and Snehota, I. (Eds) (1995), Developing Relationships in Business Networks,
Routledge, London.
Hardin, R. (1982), Collective Action, Resources for the Future, Washington, DC.
Heckathorn, D. (1996), ``The dynamics and dilemmas of collective action’’, American
Sociological Review, Vol. 61 No. 2, pp. 250-77.
Herrigel, G. (1993), ``Power and the redefinition of industrial districts: the case of BadenWurttemberg’’, in Grabher, G. (Ed.), The Embedded Firm: On the Socioeconomics of
Industrial Networks, Routledge, London.
Lorenz, E. (1992), ``Trust, community, and cooperation: toward a theory of industrial
districts’’, in Storper, M. and Scott, A. (Eds), Pathways to Industrialization and Regional
Development, Routledge, London.
Lundgren, A. (1992), ``Coordination and mobilisation processes in industrial networks’’, in
Axelsson, B. and Easton, G. (Eds), Industrial Networks: A New View of Reality,
Routledge, London.
Macneil, I. (1990), ``Political exchange as relational contract’’, in Marin, B. (Ed.), Generalised
Political Exchange: Antagonistic Cooperation and Integrated Policy Circuits, Campus
Verlag/Westview Press, Frankfurt.
Mattsson, L.-G. (1985), ``An application of a network approach to marketing: defending and
changing market positions’’, in Dholakia, N. and Arndt, J. (Eds), Changing the Course of
Marketing: Alternative Paradigms for Widening Market Theory (Supplement 2), JAI
Press, Greenwich, CT.
JO U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 2 00 1
1 63
Miller, B. (1992), ``Collective action and rational choice ± place, community and the limits to
individual self-interest’’, Economic Geography, Vol. 68 No. 1, pp. 22-42.
Munck, G. (1995), ``Actor formation, social coordination and political strategy ± some
conceptual problems in the study of social-movements’’, Sociology ± The Journal of the
British Sociological Association, Vol. 29 No. 4, pp. 667-85.
Naude , P. and Turnbull, P. (Eds) (1998), Network Dynamics in International Marketing,
Elsevier Science, Oxford.
North, D. (1991), Institutions, Institutional Change and Economic Performance, Cambridge
University Press, Cambridge.
Oliver, P. and Marwell, G. (1988), ``The paradox of group size in collective action: a theory of
the critical mass, II’’, American Sociological Review, Vol. 53, pp. 1-8.
Olson, M. (1965), The Logic of Collective Action, Harvard University Press, Cambridge, MA.
Posner, E. (1996), ``The regulation of groups ± the influence of legal and non-legal sanctions
on collective action’’, The University of Chicago Law Review, Vol. 63 No. 1, pp. 133-97.
Powell, W. (1990), ``Neither market nor hierarchy: network forms of organization’’, in Staw, B.
(Ed.), Research in Organizational Behavior, Vol. 12, JAI Press, Greenwich, CT.
Sabel, C., Herrigel, G., Kazis, R. and Deeg, R. (1987), ``How to keep mature industries
innovative’’, Technology Review, Vol. 90 No. 3, pp. 26-35.
Scott, R. (1987), Organisations: Rational, Natural and Open Systems, 2nd ed., Prentice-Hall,
Englewood Cliffs, NJ.
Taylor, M. and Singleton, S. (1993), ``The communal resource ± transaction costs and the
solution of collective action problems’’, Politics & Society, Vol. 21 No. 2, pp. 195-214.
Turnbull, P., Ford, D. and Cunningham, M. (1996), ``Interaction, relationships and networks in
business markets: an evolving perspective’’, The Journal of Business & Industrial
Marketing, Vol. 11 No. 3/4, pp. 44-62.
UdeÂhn, L. (1993), ``Twenty-five years with `The logic of collective action’’’, Acta Sociologica ,
Vol. 36 No. 3, pp. 239-61.
Van Waarden, F. (1992), ``Emergence and development of business interest associations. An
example from The Netherlands’’, Organization Studies, Vol. 13 No. 4, pp. 521-62.
&
1 64
J O U R N A L O F B U S IN E S S & IN D U S T R IA L M A R K E T IN G , V O L . 16 N O . 3 20 01